SCODEL: Shared Canonical Ontology Definitions for Food Entity Linking and Ontology-Grounded Recipe Knowledge Graphs

Nuhu Ibrahim; Rishi Ravikumar; Ayuba Dauda; Fuqi Xu; Robert Stevens; Riza Batista-Navarro

cs.AIOct 20, 2026

SCODEL: Shared Canonical Ontology Definitions for Food Entity Linking and Ontology-Grounded Recipe Knowledge Graphs

Authors: Nuhu Ibrahim, Rishi Ravikumar, Ayuba Dauda, Fuqi Xu, Robert Stevens, Riza Batista-Navarro

Organizations: Department of Computer Science, The University of Manchester, United Kingdom

Abstract

Identifying recipe entities and linking them to ontology concepts is fundamental for semantic recipe understanding. However, food mentions in recipes are often ambiguous, context-dependent, and lexically diverse, making ontology entity linking (EL) challenging. We propose SCODEL, a streamlined, model-agnostic approach based on a central idea: contextual food mentions and ontology concepts that refer to the same entity should learn the same canonical ontology-oriented definition. SCODEL uses these shared definitions as a semantic normalisation layer, providing complementary semantic information to existing EL models without requiring a new linking architecture. We further introduce ReciLinkD, the first benchmark for contextual food ontology entity linking in recipes. Experiments show that SCODEL consistently improves three modular EL models (ReFinED, GSM-EL, and BLINK): SCODEL x BLINK achieves the highest Hit@1 on seen concepts (88.33%), while SCODEL x GSM-EL achieves the highest Hit@1 on unseen concepts (83.15%), outperforming the strongest large language model (LLM) based EL baseline in each setting. Experiments on a biomedical ontology benchmark further demonstrate that SCODEL generalises beyond the food domain. Finally, we use SCODEL to ground food mentions at scale and construct an ontology-grounded recipe knowledge graph, demonstrating its utility for ontology-aware recipe retrieval and corpus-level food analysis.

Figures & tables

Figure 1: Overview of the proposed SCODEL methodology.

Method	Supervised (Seen)				Zero-Shot (Unseen)				AVG
Method	Hit@1	Hit@5	MRR	H-Acc	Hit@1	Hit@5	MRR	H-Acc	Hit@1
LLM4EL	63.29	88.33	73.90	66.89	63.72	90.21	74.92	71.05	63.51
LLMAEL	85.25	94.51	89.61	86.28	80.71	94.70	86.47	81.52	82.98
LLM Rerank@10	82.68	89.88	85.97	83.70	83.02	92.80	87.19	83.83	82.85
LLM Rerank@50	81.13	88.51	84.52	81.65	80.43	90.35	84.87	81.39	80.78
Lexical Matching	52.66	73.93	61.62	54.15	56.79	75.27	65.81	57.99	54.73

Table 1: Main entity linking results on the ReciLinkD benchmark under seen and unseen settings. SCODEL uses manually generated canonical definitions as supervision and Mistral-7B, the best-performing individual definition-learning backbone (Table 5 ). LLM4EL uses Mistral-7B for both definition generation and embedding generation, LLMAEL uses Mistral-7B for context augmentation and ReFinED for entity linking (Appendix B.4 : Table 11 ), and LLM Rerank@10 and Rerank@50 use Mistral-7B for reranking the ReFinED predictions. Best results are bold, second-best results are underlined, and AVG denotes the mean Hit@1 across both settings.

Method	Hit@1	Hit@5	MRR
ReFinED Baseline	62.75	77.36	68.96
LLM4EL	63.00	81.10	–
LLMAEL	69.97	85.73	77.08
SCODEL	72.45	87.87	79.35

Table 2: Results on the GSCplus dataset with the Human Phenotype Ontology ( Luo et al., 2021 ) using ReFinED as the downstream entity linking model for LLMAEL and SCODEL. LLM4EL results are taken directly from Xu et al. (2026) , which does not report MRR; all other results are obtained using our implementation on the same benchmark. Best results are bold and second-best results are underlined.

Training Strategy	Hit@1	MRR
ReFinED Baseline	81.82	85.42
Mention Def. Learning Only	81.65	86.85
Ontology Def. Learning Only	82.68	86.68
SCODEL	85.08	89.11

Table 3: Effect of different definition learning strategies using ReFinED as the downstream entity linking model. Best results are bold and second-best results are underlined.

Definition Generator	Hit@1	MRR
ReFinED Baseline	81.82	85.42
Frozen Mistral-7B	82.16	87.46
Frozen GPT-5.1-2025-11-13	82.50	86.93
SCODEL (Mistral-7B)	85.08	89.11

Table 4: Comparison of SCODEL with frozen definition generators that did not learn shared definitions. ReFinED is used as the downstream EL model. Best results are bold and second-best results are underlined.

Method	Def. Generator	Hit@1	MRR
Single	Llama-3-8B	82.33	86.43
	Mistral-7B	85.08	89.11
	Qwen2.5-7B	83.88	87.94
Ensemble	Hard Voting	85.42	90.16
Ensemble	Soft Voting	86.11	90.59

Table 5: Comparison of backbone models for SCODEL shared definition learning and generation. ReFinED is used as the downstream entity linking model.

Mention with Context	Gold Concept	Predicted Concept
Add the dairy milk to the sauce and stir until smooth.	animal milk ( FOODON_02010013 )	cow milk ( FOODON_03307455 )
Toss the salad with fresh snap peas before serving.	pea ( FOODON_03301300 )	sugar snap pea plant ( FOODON_03412467 )
Garnish with sliced cherry tomatoes before serving.	tomato ( FOODON_03301453 )	sweet cherry ( FOODON_03311457 )
Roast the baby potatoes until golden and tender.	potato ( FOODON_03301449 )	red bliss potato ( FOODON_03307121 )

Table 6: Examples of common entity linking errors made by the SCODEL (Mistral-7B) x ReFinED model.

Appendix figures & tables25 assets

Supplementary material from the paper’s appendix.

Appendix

Metric	Value
Unique manually selected food mentions	4,500
Unique manually linked food mentions	3,979
Unique unlinked food mentions	521
Unique FoodOn food concepts	979
Rows with synonyms	2,023
Rows with superclasses	2,970

Appendix

Table 7: Descriptive statistics for ReciLinkD.

Training Definition Source	Hit@1	MRR
ReFinED Baseline	81.82	85.42
OpenAI GPT-5.1-2025-11-13	84.91	89.06
Claude Sonnet 5	84.05	88.28
Qwen3-235B	83.36	88.07
Human Annotated	85.08	89.11

Appendix

Table 8: Effect of different sources used for constructing the canonical definitions. All experiments use Mistral-7B as the shared definition learner and generator and ReFinED as the downstream entity linking model. Best results are bold and second-best results are underlined.

Metric	Mistral-7B	SCODEL
Hit@1	27.62	46.31
Hit@5	51.63	70.84
Hit@10	64.67	78.73
MRR	39.92	57.72
Gold Sim.	0.6771	0.8312
Global Margin	0.3148	0.3595

Appendix

Table 9: Semantic alignment before and after SCODEL shared canonical definition learning. Mistral-7B denotes zero-shot definition generation using the original model as in LLM4EL Xu et al. (2026) . Best results are bold.

Model Family	Parameters	Hit@1	MRR
LLaMA-3	1B	81.82	85.60
LLaMA-3	3B	83.53	87.12
LLaMA-3	8B	82.33	86.43
LLaMA-3	70B	84.91	88.72

Appendix

Table 10: Effect of model size on SCODEL shared definition learning and generation. ReFinED is used as the downstream entity linking model. Best results are bold and second-best results are underlined.

Entity Linking Model	Hit@1	Hit@5	MRR
ReFinED	85.25	94.51	89.61
GSM-EL	81.81	93.65	87.06
BLINK	85.22	93.82	89.60

Appendix

Table 11: Performance of LLMAEL across different downstream entity linking models. Mistral-7B is used as the language model across all configurations. Best results are in bold.

Annotator Pair	F1 (%)	Cohen’s $\boldsymbol{\kappa}$
A–B	85.49	0.7178
A–C	90.56	0.7254
A–D	91.31	0.7404
B–C	89.53	0.7219
B–D	88.68	0.7203
C–D	92.43	0.7258

Appendix

Table 12: Pairwise inter-annotator agreement for the 300 ReciLinkD food mentions independently annotated by all four annotators.

Manual Evaluation Outcome	Count	Perc.
Correct and retained	11,580	57.90%
Correct unsuitable food mention	7,226	36.13%
Incorrect FoodOn prediction	2,194	10.97%

Appendix

Table 13: Manual evaluation of SCODEL

\times

ReFinED predictions for 20,000 additional ReciFine food mentions.

Source	Mention–Concept Pairs
ReciLinkD	3,979
ReciLinkD Expansion	11,580
Combined resource	15,559

Appendix

Table 14: Composition of the expanded manually verified food mention and ontology concept linking resource.

Vocabulary Expansion	Count
Original verified food mention names	15,559
FoodOn concept labels	979
Names from FoodOn concept synonyms	253
Singular/plural name variants	16,738
Final expanded vocabulary	33,529

Appendix

Table 15: Expansion of the verified food-name vocabulary used to ground additional ReciFine mentions.

Statistic	Value
ReciFine recipes considered	2,206,253
Unique food mentions matched to vocabulary	19,504
Recipes grounded to food mentions	2,062,103

Appendix

Table 16: Coverage obtained by propagating the expanded verified food mention and ontology concept vocabulary across ReciFine.

Knowledge Graph Statistic	Value
ReciFine recipes considered	2,206,253
Grounded recipe nodes	2,062,103
Grounded food mention nodes	14,177,860
Directly grounded FoodOn concepts	979
Inherited FoodOn concepts	2,209
Recipe $\rightarrow$ mention edges	14,177,860

Appendix

Table 17: Statistics of the expanded ontology-grounded ReciFine knowledge graph. Of the 2,206,253 ReciFine recipes considered, 2,062,103 contain at least one grounded food mention and are represented as recipe nodes. Derived transitive ontology closure edges are reported separately and excluded from the principal graph-edge total.

Figure 2: Prompt template for determining whether the recipe contains food mentions belonging to the queried category.

Annotator Pair	Agreement	Cohen’s $\kappa$
Human 1 – Human 2	88.95%	0.76
Human 1 – GPT-5.6-Sol	82.00%	0.64
Human 2 – GPT-5.6-Sol	80.50%	0.61

Appendix

Table 18: Agreement on 200 randomly sampled query and recipe relevance judgements. Agreement denotes the percentage of identical binary relevance labels, while Cohen’s kappa (

\kappa

) accounts for chance agreement.

Query	Lexical			Semantic			Ontology
Query	P	R	F1	P	R	F1	P	R	F1
Fruit	86.67	4.34	8.26	91.26	10.46	18.76	71.77	76.08	73.87
Vegetable	85.11	3.45	6.64	96.19	8.72	15.99	82.06	18.57	30.28
Legume	0.00	0.00	0.00	0.00	0.00	0.00	60.37	52.11	55.93
Meat	80.25	20.93	33.21	85.64	26.89	40.93	94.76	34.94	51.06
Pork	91.49	21.83	35.25	91.49	21.83	35.25	88.24	15.23	25.97

Appendix

Table 19: Comparison of lexical, semantic, and ontology-aware recipe retrieval across 15 higher-level food-category queries. P, R, and F1 denote precision, recall, and F1-score (%), respectively, over the 2,000-recipe ReciFine evaluation sample. The Average row reports averaged performance across the 15 queries. Best results are in bold.

FoodOn Category	Recipes	Prevalence
Plant food product	1,744,295	84.59%
Animal food product	1,510,986	73.27%
Vegetable food product	1,297,374	62.92%
Dairy food product	1,046,656	50.76%
Fruit	978,283	47.44%
Condiment food product	939,353	45.55%

Appendix

Table 20: Recipe-level prevalence of 20 selected higher-level FoodOn categories obtained through ontology hierarchy propagation on the 2,062,103 grounded recipes. Categories were selected for culinary interpretability, diversity, and to reduce redundancy between closely related ontology concepts.

Category A	Category B	Recipes	Proportion
Plant food product	Animal food product	1,303,624	74.74%
Animal food product	Vegetable food product	969,055	64.13%
Plant food product	Dairy food product	917,684	52.61%
Vegetable food product	Fruit	891,116	68.69%
Plant food product	Condiment food product	831,885	47.69%
Animal food product	Condiment food product	746,545	49.41%

Appendix

Table 21: Selected co-occurrence relationships between higher-level FoodOn categories. Proportion denotes the percentage of recipes containing Category A that also contain Category B.

Figure 3: Prompt template for generating shared food mention and ontology concept pair definitions.

Figure 4: Prompt template used for shared canonical definition learning and generation from contextual food mentions in SCODEL, and for definition generation from contextual food mentions in LLM4EL.

Figure 5: Prompt template used for shared canonical definition learning and generation from FoodOn concept metadata in SCODEL, and for definition generation from FoodOn concept metadata in LLM4EL.

Setting	Value
Models	GPT-5.1-2025-11-13, Claude Sonnet 5, Qwen3-235B
Input	Aligned mention–concept pair
Output	Shared canonical definition
Parameter updates	None

Appendix

Table 22: Configuration for automatic paired canonical definition generation used to construct alternative SCODEL supervision.

Setting	Value
Base models	LLaMA-3-8B, Mistral-7B, Qwen2.5-7B
Finetuning method	LoRA (Unsloth)
LoRA rank ( $r$ )	16
LoRA alpha	16
LoRA dropout	0
LoRA bias	None

Appendix

Table 23: Training configuration for SCODEL shared canonical definition learning.

Setting	Value
Decoding	Sampling
Temperature	0.7
Top- $p$	0.9
Maximum new tokens	1024

Appendix

Table 24: Inference configuration for canonical definition generation after SCODEL training.

Figure 6: Prompt template used for generating LLMAEL food mention augmentation description.

Figure 7: Prompt template used for LLM-reranking entity linking experiments.

Statistic	Train	Seen	Unseen
Food mentions	2,660	583	736
Unique ontology concepts	803	255	176
Mean mention–concept pair Jaro similarity	0.788	0.758	0.794
Exact pair match (%)	18.98	11.84	18.61
Mean food mention token length	1.81	1.83	1.79

Appendix

Table 25: Dataset split statistics and characteristics. Jaro similarity ( Jaro, 1989 ) is computed between each recipe food mention and the label of its gold FoodOn concept after text normalisation; higher values indicate greater lexical similarity.

Entity Linking

Entity linking maps mentions of people, places, products, and concepts to the corresponding entries in a knowledge base.

See all 25 papers

Beyond Fine-Tuning: Robust Food Entity Linking under Ontology Drift with FoodOntoRAG
Mar 10, 2026Jan Drole, Ana Gjorgjevikj, Barbara Korouši'c Seljak +1Entity Linking Language Model Robustness
Computer Systems Department, Jožef Stefan International Postgraduate School, Ljubljana, Slovenia · Computer Systems Department, Jožef Stefan Institute, Ljubljana, Slovenia
Epicure: Navigating the Emergent Geometry of Food Ingredient Embeddings
May 21, 2026Jakub Radzikowski, Josef ChenGraph Embedding
ReGraph: Learning to Generate Recipe Graphs from Food Images
Aug 7, 2026Guoshan Liu, Bin Zhu, Pengkun Jiao +3Graph Generation Graph Representation Learning
Institute of Trustworthy Embodied AI, Fudan University, China and Shanghai Key Laboratory of Multimodal Embodied AI, China · Singapore Management University, Singapore

Mar 10, 2026J/K move · Enter open · S save

Beyond Fine-Tuning: Robust Food Entity Linking under Ontology Drift with FoodOntoRAG

Jan Drole, Ana Gjorgjevikj, Barbara Korouši'c Seljak, Tome Eftimov

Computer Systems Department, Jožef Stefan International Postgraduate School, Ljubljana, Slovenia · Computer Systems Department, Jožef Stefan Institute, Ljubljana, Slovenia

Entity Linking Language Model Robustness Retrieval-Augmented Generation

Open paper AskPDF arXiv

Standardizing food terms from product labels and menus into ontology concepts is a prerequisite for trustworthy dietary assessment and safety reporting. The dominant approach to Named Entity Linking (NEL) in the food and nutrition domains fine-tunes Large Language Models (LLMs) on task-specific corpora. Although effective, fine-tuning incurs substantial computational cost, ties models to a particular ontology snapshot (i.e., version), and degrades under ontology drift. This paper presents FoodOntoRAG, a model- and ontology-agnostic pipeline that performs few-shot NEL by retrieving candidate entities from domain ontologies and conditioning an LLM on structured evidence (food labels, synonyms, definitions, and relations). A hybrid lexical--semantic retriever enumerates candidates; a selector agent chooses a best match with rationale; a separate scorer agent calibrates confidence; and, when confidence falls below a threshold, a synonym generator agent proposes reformulations to re-enter the loop. The pipeline approaches state-of-the-art accuracy while revealing gaps and inconsistencies in existing annotations. The design avoids fine-tuning, improves robustness to ontology evolution, and yields interpretable decisions through grounded justifications.

Abstract

Entity Linking

Explore similar work

Beyond Fine-Tuning: Robust Food Entity Linking under Ontology Drift with FoodOntoRAG

Epicure: Navigating the Emergent Geometry of Food Ingredient Embeddings

ReGraph: Learning to Generate Recipe Graphs from Food Images